Stephen Brand
www.statisticsdoc.com
Helga,
These levels of reliability are usually regarded as modest, but usually
acceptable for a research study in peer reviewed journals. One
consideration that reviewers may keep in mind is whether other scales that
have been used in the literature to measure similar constructs, if there are
any, have exhibited higher levels of internal consistency.
When alpha drops below .7, most of the variance consists of measurement
error, rather than true score variance.
With alpha levels like these, I would be a little concerned about
attenuation of effects due to measurement error. It is possible for you to
increase the internal consistency of the scales by dropping items, or
combining the items into shorter but more internally consistent scales?
HTH,
Stephen Brand
For personalized and professional consultation in statistics and research
design, visit
www.statisticsdoc.com
-----Original Message-----
From: SPSSX(r) Discussion [mailto:
[hidden email]]On Behalf Of
Helga Walz
Sent: Tuesday, November 21, 2006 6:18 PM
To:
[hidden email]
Subject: your assistance please
To the experts with thanks ahead of time,
For a research study I put together 9 items to reflect the concept
"communication gap between young and old".
Scoring is on a 5-point Likert-type scale. 56 students provided answers.
Cronbach's alpha is .755
For 2 other, different scales alpha is .767 for 7 items and .780 for 13
items respectively.
Are these alpha levels sufficient for the scales to be used in a
research project?
Thank you,
Helga Walz
Helga S. Walz, Ph.D.
Division of Applied Behavioral Sciences
University of Baltimore
1420 North Charles Street
Baltimore, Maryland 21201
Tel.: 410 837-5279
Fax.:410 837-4059
Locked
